feat: add Elfa AI provider (Auto condition engine + Iris market intelligence, x402) - #216
Open
nero-sensei wants to merge 1 commit into
Open
feat: add Elfa AI provider (Auto condition engine + Iris market intelligence, x402)#216nero-sensei wants to merge 1 commit into
nero-sensei wants to merge 1 commit into
Conversation
nero-sensei
force-pushed
the
feat/add-elfa-ai
branch
from
August 3, 2026 04:46
9712cf9 to
3990c10
Compare
nero-sensei
force-pushed
the
feat/add-elfa-ai
branch
from
August 3, 2026 04:53
3990c10 to
d09a699
Compare
nero-sensei
marked this pull request as ready for review
August 3, 2026 04:58
Greptile SummaryThe PR adds the native
Confidence Score: 5/5The PR appears safe to merge, with no concrete changed-code defect established in the provider metadata or generated API contract. The provider layout and frontmatter follow registry conventions, paid and free operations are consistently documented, and the reported catalog checks cover all 18 added endpoints without an accepted runtime, publication, or security failure. Important Files Changed
Sequence DiagramsequenceDiagram
participant Agent
participant Elfa as Elfa AI API
participant Webhook as Agent Webhook
Agent->>Elfa: Validate EQL query (free)
Elfa-->>Agent: Validation result and exact cost
Agent->>Elfa: Create query (x402 payment)
Elfa-->>Agent: Query ID and active status
loop Query lifecycle
Agent->>Elfa: Poll status or open SSE stream (free)
Elfa-->>Agent: Status and notifications
end
Elfa->>Webhook: Fire configured action when condition matches
Agent->>Elfa: Cancel query (free)
Reviews (1): Last reviewed commit: "feat: add Elfa AI provider (Auto conditi..." | Re-trigger Greptile |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds Elfa AI (
elfa-ai/api).Elfa is real-time data infrastructure for financial AI, powered by our intelligence stack Iris. Iris synthesises signals across social, on-chain and off-chain market data, then traces their second and third-order effects, so agents act on greater context.
These endpoints serve the same signals our own finance agent runs on, pay-per-call over x402 — no API key and no account. We run them ourselves at
https://api.elfa.ai, no gateway in front.Auto — the condition engine
The part we expect agents to reach for most. Describe what to watch and Elfa fires a webhook when it happens: register an EQL query against price or social conditions, and it is evaluated continuously. Identity is
SHA256(x-elfa-agent-secret), so an agent invents a secret once and reuses it to reach its own queries.Creating a query is the only paid step — validating, polling, cancelling, listing LLM sessions and the SSE notification stream are free.
Iris market intelligence
Point-in-time reads of what the market is saying and doing: trending assets and contract addresses surfacing on X and Telegram, smart-account stats, keyword mention feeds, token news, event summaries and narratives, plus an LLM chat endpoint that interprets them.
Payment
solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp), plus Base, Arbitrum, Polygon and Avalanche — every rail is quoted in the same 402, so the caller pays on whichever they already hold.exactscheme, challenge in thePAYMENT-REQUIREDheader.Endpoints
18 total — twelve paid, six free.
speedspeedNotes for reviewers
Auto query pricing is dynamic.
POST /x402/v2/auto/queriescosts a $0.045 baseline plus each simulated LLM call the query needs ($0.045 fast, $0.162 expert). The listed variants describe the common single-call case rather than a floor or ceiling.PAY.mdsays so and points callers at the freevalidateendpoint, which returns the exact cost before anything is created. Happy to express this differently if you would rather see the baseline advertised.The six free Auto endpoints carry explicit
price_usd: 0variants. They are identity-scoped on anx-elfa-agent-secretheader, so an unauthenticated probe cannot classify them. The zero-priced variants are what let them publish honestly instead of being dropped.The spec is generated, not hand-maintained.
openapi.jsonis produced from the same price table our payment middleware charges from, so the advertised price cannot drift from what we bill. Refreshing it is a single command on our side.Checks
pay catalog check providers/elfa/api/PAY.md --strictis also clean, with zero static warnings.